Particulate Matter 2.5 As A Contributing Factor To COVID-19 Events In Washington State: A Supervised Learning Approach: A study of the correlation between PM 2.5, COVID-19 positive case, hospitalization and death counts, and other environmental and socio-economic variables based on state and county levels.
Translated title
Particulate Matter 2.5 As A Contributing Factor To COVID-19 Events In Washington State: A Supervised Learning Approach
Author
Lenihan-Clarke, Marie Ann
Term
4. term
Publication year
2021
Pages
38
Abstract
Luftforurening er forbundet med flere hjerte- og luftvejssygdomme. Mindre vides om, hvordan dag-til-dag-ændringer i fine partikler (PM2,5) — partikler mindre end 2,5 mikrometer — hænger sammen med COVID-19-smittetal, indlæggelser, dødsfald og sygdomsforløb. Fordi PM2,5 er meget små, kan de passere kroppens naturlige forsvar, komme ind i blodbanen og beskadige celler, hvilket kan øge modtageligheden for virusinfektioner. Dette studie gennemfører en kortsigtet tidsserieanalyse på stats- og amtsniveau for at undersøge, hvordan PM2,5-niveauer relaterer til COVID-19-hændelsesrater. Vi kombinerer luftkvalitetsdata fra Environmental Protection Agency (EPA)-godkendte målestationer med frivilligt indsamlede sensordata (Volunteered Geographic Information, VGI) og sammenligner deres pålidelighed og konsistens. Med statistiske analyser og visualiseringer vurderer vi sammenhænge mellem PM2,5 og COVID-19-indikatorer. Vi finder, at langsigtede gennemsnit af PM2,5 har en tydeligere sammenhæng med samlede positive COVID-19-tilfælde end daglige udsving. Samtidig er daglige COVID-19-opgørelser nyttige til at træne og teste maskinlæringsmodeller, der forudsiger smitteforekomst ud fra PM2,5 og andre variable i studiet.
Air pollution is linked to higher rates of heart and respiratory diseases, but less is known about how day-to-day changes in fine particulate matter (PM2.5)—particles smaller than 2.5 micrometers—relate to COVID-19 case counts, hospitalizations, deaths, and recovery times. Because PM2.5 is very small, it can bypass the body’s defenses, enter the bloodstream, and damage cells, potentially increasing susceptibility to viral infections. This study conducts a short-term time-series analysis at state and county levels to examine how PM2.5 levels align with COVID-19 event rates. We combine air quality data from Environmental Protection Agency (EPA)-approved monitors with community-contributed sensor data (Volunteered Geographic Information, VGI) and compare their reliability and consistency. Using statistical tests and data visualizations, we assess correlations between PM2.5 and COVID-19 indicators. We find that long-term averages of PM2.5 show a clearer relationship with cumulative positive COVID-19 cases than day-to-day fluctuations. At the same time, daily COVID-19 records are useful for training and testing machine-learning models that predict case occurrence using PM2.5 and other variables included in the study.
[This summary has been rewritten with the help of AI based on the project's original abstract]
Keywords
Documents